Is it time to revise RECIST criteria? One radiologist thinks so

The causes of renal cell carcinoma progression may be more complex than outlined in the Response Evaluation Criteria in Solid Tumors version 1.1 (RECIST 1.1), according to findings published May 14 in Radiology.

One radiologist in particular, Christiane K. Kuhl, of University Hospital Aachen’s Department of Diagnostic and Interventional Radiology in Germany, suggested the guidelines need a revision, calling the May 14 study a “wake-up call” for radiologists.

Progression-free survival (PFS)—as determined by RECIST 1.1—remains the reference standard for evaluating the effectiveness of therapy and cancer treatments, but some are concerned the criteria may not apply to patients treated with certain therapies, wrote lead author Heidi J. Coy, of Ronald Reagan UCLA Medical Center in Los Angeles, and colleagues.

“Prior studies have attempted to redefine response assessment in patients treated with targeted cancer agents due to concerns that RECIST 1.1 may result in a different response pattern, as the mode of action in targeted cancer agents differs from that of cytotoxic therapy in which RECIST 1.1 was initially validated," Coy et al. added.

"Our hypothesis was that progression based on nonmeasurable disease, as defined by RECIST 1.1, is more common than the RECIST 1.1 guidelines would suggest.”

The researchers conducted a secondary analysis of 395 patients included in the phase III METEOR trial conducted between 2013 and 2014; it included patients (mean age 61) with metastatic clear cell renal cell carcinoma, with at least one targeted lesion at baseline and one follow-up time point who had progressive disease based on RECIST 1.1. Each time point consisted of a chest, abdominal and pelvic scan.

Overall, 73 (18.5%) patients had progression due to non-target disease, 105 (26.6%) had new lesions and 126 (31.9%) had progression of target lesions.

Patients whose progression was detected because of changes in non-target disease (median PFS of 2.8 months) or to new lesions (median 3.6 months) have a shorter PFS than those whom progression was defined by an increase in the sum of diameters of target lesions (median 5.4 months).

Additionally, the most common site of a new metastatic lesion was bone, lymph node, liver and lung, in that order.

“In conclusion, although the rules governing target and new lesion evaluation are well defined, current RECIST 1.1 guidelines governing non–target disease progression leave room for reader variability and subjective interpretation,” the authors wrote. “Our current study underscores the importance of clearly defining criteria for defining non–target disease progression for future response assessments, especially as the field of targeted therapy continues to expand into immunotherapy.”

In a related editorial, Kuhl, who said the RECIST findings should be a “wake-up call” also suggested the incongruity may be because the RECIST consortium did not use prospective clinical trial data, but rather data from a warehouse of various different trials.

“In other words, the RECIST criteria are determined and refined by ‘what if’ analyses on existing data on response evaluations stored in the data warehouse,” Kuhl added. “This is somewhat surprising given the usual policy in the field of oncology, where evidence from prospective clinical trials is required to determine all aspects of treatment and where the respective outcome variables and types of analyses that are planned on the data to be collected must be laid down in writing well ahead of the actual start of data accrual.”

While RECIST appears to reduce a radiologists workload, Kuhl wrote, radiologists must measure all deposits to get a clearer picture of progressive disease outside of target lesions.

“Radiologists should be aware of this and get engaged,” Kuhl wrote. “It is our duty to ensure that our results are appropriately used to guide patient care. Let us take the lead and conduct adequately designed prospective clinical studies to investigate the appropriate use of information provided by imaging for response assessment.”